A catalogue of faint local radio AGN and the properties of their host galaxies

TL;DR

This paper catalogs faint local radio AGN by identifying galaxies with radio luminosities exceeding expectations from their star formation rates, revealing their significant contribution to low-luminosity radio emission and their host galaxy properties.

Contribution

It introduces a method to identify faint radio AGN using optical spectro-photometry and radio data, expanding the understanding of AGN populations and their host galaxies.

Findings

Faint AGN contribute significantly to the low-luminosity radio luminosity function.

Host galaxies are predominantly high-mass, early-type, and found in denser environments.

Radio emission alone is unreliable for estimating star formation rates due to faint AGN contamination.

Abstract

We present a catalogue of local ($z<0.1$) galaxies that contain faint AGN. We select these objects by identifying galaxies that exhibit a significant excess in their radio luminosities, compared to what is expected from the observed levels of star-formation activity in these systems. This is achieved by comparing the optical (spectroscopic) star formation rate (SFR) to the 1.4~GHz luminosity measured from the FIRST survey. The majority of the AGN identified in this study are fainter than those in previous work, such as in the Best and Heckman (2012) catalogue. We show that these faint AGN make a non-negligible contribution to the radio luminosity function at low luminosities (below $10^{22.5}$ W Hz$^{-1}$), and host $\sim$13 per cent of the local radio luminosity budget. Their host galaxies are predominantly high stellar-mass systems (with a median stellar mass of $10^{11}\textrm{M}_{\odot}$), are found across a range of environments (but typically in denser environments than star-forming galaxies) and have early-type morphologies. This study demonstrates a general technique to identify AGN in galaxy populations where reliable optical SFRs can be extracted using spectro-photometry and where radio data are also available so that a radio excess can be measured. Our results also demonstrate that it is unsafe to infer SFRs from radio emission \textit{alone}, even if bright AGN have been excluded from a sample, since there is a significant population of faint radio AGN which may contaminate the radio-derived SFRs.